In the United States, adolescent girls in particular are at especially heightened risk for contracting STIs (CDC, 2003), but relatively little is known about the psychological antecedents of girls'risk and protection practices, including condom use, and how these practices develop and change over time. The salience of body image among adolescent girls (Adams et al., 1993) and observed connections between body image and condom use among adult women (Schooler et al., 2005) suggest that body image should be incorporated into developmental models of adolescent girls'protection and risk behavior. Currently, no such developmental models exist. The specific aims of the proposed project are 1) to examine the nature of the association between body image and risk and protection behaviors among early adolescent girls (13-14), middle adolescent girls (15-16) and late adolescents girls (17-18), and 2) to examine the specific mechanisms by which body image contributes to risk and protection behaviors. The study will utilize a secondary data-analysis of two existing longitudinal datasets (n=508), which combine survey and interview data. Structural equation modeling, regression, and latent growth curve modeling will be used to evaluate quantitative models of risk and protection behaviors longitudinally across adolescence. Additionally, a focused qualitative analysis will explore how adolescent girls talk about risk, protection, and decision-making, illuminating mechanisms of the association between body image and risk and protection behaviors not evaluated in the quantitative analyses. PUBLIC HEALTH RELEVANCE The public health community has voiced its concern over the incidence of STIs, including HIV, among adolescents by ranking the increase of condom use among its top priorities for the decade (DHHS, 2000). It is expected that the proposed study will provide empirically-based suggestions for developing programs and interventions that improve body image, reduce risk behaviors, and increase the use of protection among adolescent girls. Findings will guide educators, health-care providers and parents wishing to promote behavioral change among girls, by identifying the optimal developmental stage for intervention and also the girls for whom intervention may be most important.